The purpose of the study is to compare the relative effects of four therapeutic modalities in treating heroin addicts. 613 heroin addicts in detoxification treatment are included in the study, 486 of whom were randomly assigned to one of three quite different therapeutic communities (varied in terms of theoretical orientation, degree of patient control, cost, etc.) or a methadone maintenance program. All patients were followed up by mailed questionnaire at six months (96% return), twelve months (94% return), and by personal interview at two years after admission (96% completed). Our sample includes 211 patients who experienced no treatment beyond detoxification who serve as a quasi-control group. Statistical analyses are described which will answer initial questions such as: which types of patients enter and are retained in each treatment modality, which factors predict social outcomes two years after admission, what is the relationship of time in treatment to outcome for the therapeutic community patients, what is the relative effectiveness of the treatment modalities, what is the validity and reliability of the data, what are the relationships between the principal outcome variables and how are the follow-up processes related to successful location?